文档介绍：毕业论文(设计)
题目基于神经网络的变压器故障检测
姓名杨文学号 0817014004
所在院(系) 电气工程学院
专业班级自控081班
指导教师侯波
完成地点陕西理工学院(北区)501楼
2012年 5 月20日
基于神经网络的变压器故障检测
杨文
(陕西理工学院电气工程学院自动化专业081班,陕西汉中 723003)
指导教师:侯波
[摘要]:电力变压器作为电力系统中最为重要的设备之一,对电力系统安全、可靠、优质、经济的运行起着决定性作用,因而,必须尽量减少变压器故障的产生。电力变压器故障检测对电力系统的经济安全运行有着重要的意义。油中溶解气体法,是最有效的发现和检测变压器故障的方法之一。神经网络对外界具有很强的模式识别分类能力和联想记忆能力,因此神经网络可以用于变压器故障检测。基于神经网络的以变压器油中溶解气体为特征量的故障检测方法为变压器故障检测提供了新的途径。
本文将采用三种不同的神经网络(BP网络、RBF网络、支持向量机)应用于变压器故障检测中,分别介绍这几种网络的基本结构和原理,并进行模型设计和仿真。
[关键词]:变压器故障检测神经网络 BP算法 RBF算法支持向量机
Based on work of transformer fault detection

Author:Yang wen
(Grade 08, Class 01,Major Automation,Department of Electrical Engineering ,Shaanxi University of Technology ,Hanzhong ,723003,Shaanxi )
Tutor :Hou Bo
Abstract : as the most important part of the power system equipment,the power transformer to the safety of the electricity system, reliable and high quality, and the operation of the economy plays a decisive role, therefore, we must try to reduce the of transformer faults. Power transformer of electric power system fault detection of the economic security has important significances. The dissolved gas method, is one the most effective and found that one of the ways to detect transformer faults. work has a strong pattern recognition classification ability and associative memory ability to the outside world, so work can be used for the transformer fault detection. Based on work to gases dissolved in transformer oil for the characteristic features of fault detection method for transformer fault detection offers a new way. Therefore.
This article will use three different work (work, work, support vector machine) used in transformer fault detection, are introduced the basic structure of work and the principle and design and simulation model.
key words : transformer ,fault detection ,work ,BP algorithm ,RBF algorithm ,support vector machine.
目录
1 绪论 1
1
1
1
3
4
2 基于神经